Abstract
Adapting step-distilled text-to-image (T2I) models through post-training incurs additional computational costs and affects native few-step generation behavior. This motivates a complementary route beyond style-specific adaptation: drawing on the visual knowledge already encoded in step-distilled T2I models to elicit stylistic capabilities through language. Pursuing this direction requires textual guidance that captures how visual attributes jointly define a style and remain applicable as the depicted content changes. To explore this approach, we introduce StyleForge, a fully automatic, training-free framework that expresses reference styles as reusable rendering instructions. By integrating overall rendering characteristics with local color and lighting behavior, StyleForge organizes visual evidence from reference images into a coherent specification of how the target style should be expressed. The specification is then compiled into textual guidance that can be reused across content prompts, enabling frozen step-distilled T2I models to render different subjects and scenes in the reference style while retaining native few-step generation. Extensive experiments show relative gains of up to 29.47\% in generation quality scores over the strongest baseline, while Pareto analysis indicates that improved stylization is accompanied by strong adherence to the requested content.
Keywords
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Sun, S., Li, Y., Lian, Y., Li, X., Zhou, X., Tian, A., Wang, Z., Li, H., Cui, Z., & Ma, C. (2026). Salvation Lies Within: Eliciting Inherent Style Transfer in Step-Distilled Diffusion Models. https://omanscience.com/en/articles/salvation-lies-within-eliciting-inherent-style-transfer-in-step-distilled-diffusion-models
MLA 9
Sun, Shengyin, et al. "Salvation Lies Within: Eliciting Inherent Style Transfer in Step-Distilled Diffusion Models." https://omanscience.com/en/articles/salvation-lies-within-eliciting-inherent-style-transfer-in-step-distilled-diffusion-models.
Chicago (author–date)
Sun, Shengyin, Yiming Li, Yingzhao Lian, Xing Li, Xingzhi Zhou, Anxin Tian, Zhili Wang, Haoyang Li, Ziqiang Cui, and Chen Ma. 2026. "Salvation Lies Within: Eliciting Inherent Style Transfer in Step-Distilled Diffusion Models." https://omanscience.com/en/articles/salvation-lies-within-eliciting-inherent-style-transfer-in-step-distilled-diffusion-models.
Harvard
Sun, S., Li, Y., Lian, Y., Li, X., Zhou, X., Tian, A., Wang, Z., Li, H., Cui, Z. and Ma, C. (2026) 'Salvation Lies Within: Eliciting Inherent Style Transfer in Step-Distilled Diffusion Models', Available at: https://omanscience.com/en/articles/salvation-lies-within-eliciting-inherent-style-transfer-in-step-distilled-diffusion-models.
Vancouver
Sun S, Li Y, Lian Y, Li X, Zhou X, Tian A, et al. Salvation Lies Within: Eliciting Inherent Style Transfer in Step-Distilled Diffusion Models. https://omanscience.com/en/articles/salvation-lies-within-eliciting-inherent-style-transfer-in-step-distilled-diffusion-models
IEEE
S. Sun, Y. Li, Y. Lian, X. Li, X. Zhou, A. Tian, Z. Wang, H. Li, Z. Cui, and C. Ma, "Salvation Lies Within: Eliciting Inherent Style Transfer in Step-Distilled Diffusion Models," https://omanscience.com/en/articles/salvation-lies-within-eliciting-inherent-style-transfer-in-step-distilled-diffusion-models.